Vietnam
veterans: Misrepresented and misunderstood
An empirical look at whether vets are better or worse off than their non-veteran
counterparts
Brian Carroll
Presented
to the Southern Association for Public Opinion Research
20th Annual Conference
October 3, 2001
Raleigh, N.C. (North Carolina State University)
Abstract:
The public’s image of the Vietnam veteran is largely unflattering, and the media have contributed to this image. Many see the vet as an unemployed, wife-beating alcoholic riddled with psychological problems such as clinical depression and Rambo-like fits of rage. Heart-rending anecdotes and Hollywood film treatments aside, are Vietnam veterans as a group actually worse off than their counterparts who did not serve in the war? Have we been duped into believing that the war’s effects on soldiers were almost uniformly negative and long-lasting? Since the media’s coverage of the war’s effects has focused on individuals, often relying on anecdotal accounts offered either by veterans or those claiming to be veterans, the public has had little exposure to empirical treatments of the question.
This paper is an attempt to fill the void, and it is a logical one. The data necessary to look at the question has been readily available all along. Using statistical methods and accessing an annual demographic survey conducted by the federal government, we can provide at least a basic answer to the broad question, “Are Vietnam veterans as a group actually worse off than their counterparts who did not serve?”
Common depictions in media of the Vietnam War veteran are of a down-on-his-luck social misfit who has been traumatized and permanently scarred, physically and/or emotionally. More than 25 years after the conflict’s inconclusive end, a growing number of journalists, researchers and scholars are criticizing these types of stereotypes. Writing about former President Bill Clinton’s visit to Vietnam in 2000, for example, USA Today’s Dave Moniz wrote that the evidence of recent research shows that, “most veterans are happy, well-adjusted and successful as non-veterans. Many went on to become corporate chief executives, famous authors, military leaders and candidates for the presidency” (p. A8). The image of the Vietnam vet as dysfunctional “is a myth,” Moniz quoted Steve Maxner, a Texas Tech University historian who compiled oral histories of 75 combat soldiers.
Around anniversaries and memorials of the war, media coverage often takes the perspective of an individual who experienced the war. A prototypical example is an article this year in U.S. News & World Report on how the “healing process is far from done” (Spake, 2000). From the story of a single soldier, Perry Buck, the reporter generalizes about the fortunes of veterans since the war, focusing on hepatitis C cases and “psychological fallout” (p.39). The article repeats the now famous myth that vets were spat on upon their return from the theater, a scenario sociologist Jerry Lembcke found to be merely a myth. No such incident has been documented, according Lembcke (1997).
In a front page story in March 1981, the Washington Post focused on the effects of the war on female Vietnam veterans, dramatizing the effects by telling the story of one vet, Kathy Gunson, who “could not talk about Vietnam, not for years” (p. A1). Gunson was meant to represent the 7,484 women stationed in Vietnam, 6,250 of whom were nurses. “Volumes have been written about the problems of the 2.8 million male Vietnam veterans, but no comprehensive study of female veterans has ever been done,” according to the article.
Many anecdotal tales of the horrors of fighting and of attempts to resume normal life have been proven false. This has produced literature reacting to those frauds and examining why reporters are so easily duped (Bowden, 2000). Much of this literature focus on the efforts of people like B.G. Burkett, a vet himself determined to reveal the fraudulent claims of war service and unmask the gullibility of the news media, academics, politicians and even the Veterans Administration (Roberson, 1999). A book by Burkett, in fact, tells the story of many Vietnam vets who, like the majority of those who served, “returned to the United States to be productive members of society” (p. 45). Burkett’s crusade in exposing fraudulent Vietnam War heroes is celebrated in the October 2000 issue of Brill’s Content in an article highlighting the media’s complicity in supporting the vet-as-spat-upon-outcast stereotype. The bogus war vets not only steal and dilute the real soldiers’ valor, but they have “helped plant the image of the sad-sack, haunted, violent, drug- and alcohol-addicted, wife-beating, convenience-store-robbing Vietnam veteran so deep in the American psyche that Burkett fears it cannot be uprooted” [Bowden, p. 95].
Much of the empirical data available on Vietnam veterans has to do with psychological health and post-traumatic stress disorder. Unfortunately, however, these studies look almost exclusively at Vietnam vets without comparisons to the greater population. The definitive study of post-traumatic stress disorder among Vietnam veterans, for example, was the National Vietnam Veterans Readjustment Study, a $9 million, four-year study of the 3.14 million people who served in the Southeast Asian theater (Stuttaford, 1990). The study showed that 829,000 (about 26%) of those who served have suffered from PTSD. An often-cited study conducted by the American Psychiatry Assn. showed that among Vietnam veterans, the strongest effects on homelessness were post-military psychiatric disorder, psychiatric disorder and substance abuse (Rosenheck, R. and Fontana, A., 1994). But the study didn’t compare the state of vets with non-vets.
And a 1997 study linked high rates of alcoholism and post-traumatic stress disorder among war veterans, with Vietnam vets described as those “hit the hardest,” [Danitz, 1997, p. 14]. The article makes the good point that because of medical advances, 300% more lives were saved in Vietnam than in previous wars. But many of those who survived “had to deal with severe mental trauma which drove tens of thousands to alcohol and drugs to dull the pain” [Danitz, 1997, p. 14].
The literature shows significant and widespread challenges faced by veterans of the Vietnam War, and this paper is not an attempt to minimize them. It is, however, a responsible exploration into the question of whether veterans are better or worse off in commonly used terms than their non-veteran counterparts. It is also an attempt to demonstrate that the empirical resources necessary to answer the question have been at reporters’ disposal since the war, calling into question the almost exclusive reliance by reporters on anecdotal treatments.
The March 2000 Current Population Survey is conducted by the Bureau of the Census for the Bureau of Labor Statistics. The March CPS is also referred to as the Annual Demographic File since it contains, in addition to the basic monthly demographic and labor force information, annual data on work experience, income and benefits.
The CPS sample of 65,000 households is based on the civilian, non-institutional population of the United States. They are located in 792 sample areas comprising 2,007 counties and independent cities with coverage in every state and in the District of Columbia. Of the 65,000 households, approximately 13,000 refuse to participate and are identified as “non-interview” households. The participating 52,000 households contain approximately 94,000 persons 15 years old and over, who are interviewed, and approximately 28,000 children under the age of 15, who are not interviewed. The households are “scientifically selected” on the basis of residence to represent the nation as a whole, individual states and other specified areas (CPS Survey, p. 2-1). The March CPS, which was used for this study, is supplemented by 3,500 Hispanic households, 3,000 of which are interviewed.
Of the 65,000 households in the survey, for the purposes of comparing and contrasting Vietnam veterans with non-Vietnam veterans, only “person level” records were used for persons 45 years old or older. To have served in the war, a person would have had to have been at least 18 in 1973, the war’s final year. An 18-year-old in 1973 would be 45 in 2000. Person level records in the March 2000 CPS total 133,710 before reducing for age.
The Census Bureau uses 90% confidence intervals and 0.10 levels of significance to determine statistical validity. To compile the data, each household is interviewed once a month for four consecutive months one year, and again for the corresponding time period a year later.
A few crosstabulation operations pointed to correlations between some basic dependent variables, such as earnings and education, and our independent variable, Vietnam veteran-status. First, earnings was examined for its correlation with Vietnam veteran-status. To run the crosstabs, the CPS variable WSAL_VAL, which is total wage and salary earnings, was recoded into SALWAG to reorganize continuous data as ordinal data. The recode produced the following categories:
1=$0 to $20,000
2=$20,001 to $40,000
3=$40,001 to $60,000
4=$60,001 to $80,000
5=$80,001 and up
0 was designated as missing
The CPS variable A-VET, which is veteran status that accounts for all war actions, also was recoded, creating new variable NAMVET, with nominal data left nominal but values changed to reflect only whether or not the person participated in the Vietnam War. New values are: 1=Vietnam veteran and 2=non-Vietnam veteran, even if the person is a veteran of other wars. Response 0 (children of either) was designated as missing. Here are the crosstab results:
The crosstabulated comparison of Vietnam veteran status and total wages/earnings shows that the two are strongly associated (gamma of -.38 for men and .59 for women). Veterans appear to fare better in income categories 2 through 5, or all the income categories above $20,000, and it appears even more so for female veterans than for male veterans. The percentage of male Vietnam veterans in the lowest income category is almost half that of non-Vietnam veterans, putting proportionally higher numbers of veterans than non-veterans in every other category. The biggest gaps are in the $40,001 to $60,000 and $60,001 to $80,000 wage/salary brackets, for both men and women. With a chi-square value of 933 at a d.f. of 4, the correlation is deemed comfortably significant.
Muddling the picture somewhat is the significantly greater percentage of Vietnam veterans in the $20,001 to $40,000 income category, which covers upper-low to middle income classes. A closer look within this category could show where vets are clustered compared to non-vets, at the low end or high, so the category was recoded into more narrowly defined annual wages/earnings categories:
A closer inspection at the first recode’s second-lowest income category shows that within it, Vietnam veterans still appear to earn more in wages and salaries than do their non-veteran counterparts, the same result found for the larger income comparison. As with the larger wages/earnings comparison, non-vets are proportionally higher represented in the poorest category than are Vietnam vets, 30% to 23% for males. Also, the upper end is more highly represented by male vets, 28% to 23%. The results are significant, according to the chi-square test results, and the correlation is at least moderate at -.12 for males and -.10 for females.
One additional layered crosstab operation was performed to rule out race as a confounding variable since blacks were over-represented in the war. The results dramatically showed that race was not a confounding variable. Except for the lowest income bracket, Vietnam vets of all races earned two to three times their non-vet counterparts. For American Indians/Aleut Eskimos and Asians (Race categories 3 and 4, respectively), however, the numbers in the highest income categories are too small for meaningful comparisons.
A comparison in education levels is logical, since with income the most commonly linked variable in socio-economic analyses is level of education attained. To compare education with Vietnam veteran status, the CPS variable HGA (educational attainment) was recoded as EDUC. Values were changed for more readily accessible comparisons so that 1=high school grad or less; 2=college or college graduate; and 3=graduate school through Ph.D. The analysis found:
The correlations, which are statistically significant (524 at a df of 2), indicate that veterans are more educated than their non-veteran counterparts, and do so dramatically. Nearly 56% of non-vet 45-year-olds and older in the survey are high school graduates or less, while only 36% of Vietnam vets fit this description. And 53% of vets have some college or a college degree, compared with just 35% of the non-vets 45 and up – a flip-flop of the lower educated category. For education levels, these are wide gulfs. The test for gamma shows evidence of a strong association between vet status and education at -.30, meaning that by knowing a person’s veteran status would improve an estimate of that person’s educational level attained by about 30%.
Vietnam veterans appear to be generally better educated than their non-veteran counterparts. To get a bit more clarity, a layered crosstabulation was run looking at education and earnings at the same time, since there usually is a high correlation between the level of education achieved and income levels. Only males were compared in this way. The results (next page):
The layered crosstab reveals that with no exceptions, Vietnam vets fared better than their non-veteran counterparts, meaning that in all but the lowest income category for each education level, Vietnam veterans were proportionally more highly represented. The strength of the associations were moderate to very strong, and all results were statistically significant.
A comparison of the means for annual wages/earnings for vets and non-vets, when held constant for gender, shows a male vet mean of $48,223 that is 5.2% higher than the $45,858 for male non-vets. The difference for females is more dramatic, with the mean wages/earnings figure for female vets 53% higher than that for female non-vets (nearly $40,000 for women vets and about $27,000 for non-vets). Note also, however, the large standard deviations for both veteran categories and both genders.
To get a picture of occupational status, the CPS variable MJOCC (major occupation), a very detailed and sub-classified variable, was recoded to OCCUP, which summarizes the MJOCC occupations. In the recoding, people not in the survey universe were dropped. The results are as follows, shown with a frequency chart of occupation categories:
The results are a bit confusing, showing a higher percentage of non-vets holding managerial and professional positions than veterans. This occupation category would, generally, include the higher-paying positions in most cases. Conversely, a higher percentage of veterans than non-vets were found in the service and operator/laborer categories, which generally pay less. Perhaps those vets who do have managerial/professional occupations were paid significantly more than their non-vet counterparts to produce the wages/salary advantages examined earlier. The survey data ca not supply evidence to either confirm or deny this possibility.
Another measure of income is PTOT-R, a CPS recode of total personal income. This variable was recoded for our purposes to match up in terms of categories with SALWAG (see p. 2 of this paper). In a crosstab with veteran status, with gender held constant, here is what SPSS generated:
The personal income results, especially for females (-.56), confirm what was found when analyzing wages and salary. Vietnam vets appear to earn more regardless of where the income comes from (jobs versus the stock market, for example) than do their non-vet counterparts, especially in the top three income categories. And female veterans have an even more conspicuous personal income advantage over their non-veteran counterparts.
Comparing the means of personal income and again holding gender constant also points to an economic advantage in being a Vietnam veteran. Here are the SPSS results:
(According to the CPS codebook, category 18.30 = $42,500-44,999; category 14.70 = $32,500-34,999; category 14.63 = $32,500-34,999; and category 8.49 = $17,500-19,999.)
It might be more helpful to look at employment histories. To do this, the CPS employment variable PEMLR (labor force) was recoded to discretely show who was employed, who wasn’t and who wasn’t in the labor pool at all. Our age recode also was layered to discount 75-year-olds and older, so we are comparing like-aged vets and non-vets. Our results show vets with a slightly lower unemployment rate and, generally, longer work lives. Here are the full SPSS-generated results, with chi-square for significance and gamma for strength of association:
Since insurance status also is an indicator of general well-being, a crosstab comparison of vets and non-vets was run looking at levels of insured under Medicare, Medicaid, CHAMPUS and private insurance programs. Surprisingly, a higher percentage of vets were covered by private programs than non-vets, while non-vets much more often turned to Medicare and Medicaid. More than 82% of vets were covered by private programs, which generally are considered to be superior in terms of breadth of coverage to either Medicaid and Medicare. This compares to about 71% of non-vets covered by private programs. The results for private insurance levels are presented first, followed by Medicare, Medicaid and CHAMPUS, with chi-square results for each. HI_YN is the CPS variable for private insurance programs. For all four, 1=YES (covered) and 2=NO (not covered).
The question under examination here was, “Are Vietnam veterans as a group actually worse off than their counterparts who did not serve in that war?” Or have we been duped into believing that the war’s effects on soldiers were almost uniformly negative and long-lasting? From the comparisons drawn from the March 2000 CPS, it would seem many have been duped. Vietnam vets fare better than do their non-vet counterparts in the major demographic categories of wages/earnings, personal income and education. They do not show higher rates of unemployment and, generally, report slightly better health than do their non-veteran counter-parts.
Many questions remain. How do Vietnam vets compare with vets of other wars, for example? Though this study showed they do better than their non-vet counterparts, they probably aren’t as far ahead as WWII vets were compared to their counterparts. The Vietnam War was different in significant ways, with alcoholism, drug addiction and post-traumatic stress disorder rates all much higher (see studies reviewed above). Finer distinctions drawn between men and women, whites and minorities, officers and enlisted, for instance, also would prove valuable. There is still much to do.
Veteran Status (with pie chart)
Bowden, M. (2000). Lies and more lies: The militia and the media, part I. Brill's Content 3: 95-139.
Burkett, B.G. and Whitley, G. (1998). Stolen Valor: How the Vietnam Generation was Robbed of Its Heroes and Its History. Verity Press: Dallas, Texas.
Danitz, T. (1997). Drowning the demons of war. Insight on the News 13(8): 14-15.
Earley, P. (March 25, 1981). Effects of War on Female Vietnam Veterans Are Only Now Emerging; The Forgotten Women of the Vietnam War. Washington Post, A1.
Lembcke, J. (1997) The Spitting Image: Myth, Memory, and the Legacy of Vietnam. NYU Press: New York City.
Rosenheck, R. and Fontana, A. (1994). A model of homelessness among male veterans of the Vietnam War generation. American Journal of Psychiatry (151), 3, 421-428.
Ryan, M. (1997). “Robert Olen Butler's Vietnam veterans: strangers in an alien home.” The Midwest Quarterly 38(3): 274-93.
Spake, A. (May 1, 2000). The healing process is far from done: health problems of Vietnam veterans. U.S. News & World Report (128), 17, 39.
Stuttaford, G. (March 9, 1990). Trauma and the Vietnam War Generation: Report of the Findings from the National Vietnam Veterans Readjustment Study. Publishers Weekly (237), 10, 57-68.
Ulin, D. (July 3, 1998). Were Vietnam Vets Spat Upon When They Returned Home, Or Is That an Urban Myth? Los Angeles Times, E1.